Energy policy on shaky ground? : A study of CCS scenarios

(1)

Energy policy on shaky ground? A study of

CCS scenarios

Mårten Bryngelsson and Anders Hansson

Linköping University Post Print

N.B.: When citing this work, cite the original article.

Original Publication:

Mårten Bryngelsson and Anders Hansson, Energy policy on shaky ground? A study of CCS

scenarios, 2009, Energy Procedia, (1), 1, 4673-4680.

http://dx.doi.org/10.1016/j.egypro.2009.02.290

Copyright: Elsevier

http://www.elsevier.com/

Postprint available at: Linköping University Electronic Press

http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-19317

(2)

Energy

Procedia

Energy Procedia 00 (2008) 000–000

www.elsevier.com/locate/XXX

GHGT-9

Energy policy on shaky ground? A study of CCS-scenarios

Mårten Bryngelsson

a,

* and Anders Hansson

a, b, c

a_{KTH, School of Chemical Sciences, Teknikringen 50, SE-10044, Stockholm} b_{Linköping University, Centre for Climate Science and Policy Research, SE-60174, Norrköping}

c_{Linköping University, Department of Technology and Social Change, SE-58183, Linköping}

Elsevier use only: Received date here; revised date here; accepted date here

Abstract

Scenarios play an important role for the societal acceptance of CCS. This paper looks into influential reports containing CCS scenarios and analyses results, key assumptions and drivers for CCS’ deployment. Significant uncertainties regarding CCS’ development were in several cases excluded or marginalized. Despite these shortcomings, scenarios support a massive deployment of CCS and reflect an undivided optimism. If CCS would fail to meet the high expectations a backlash could follow. Indications were found that new scenarios including uncertainties are needed to balance this over-optimism. So-called unpleasant scenarios are often valuable in helping decision makers develop flexible strategies and policies.

Keywords: CCS; economic modelling; scenario studies; policy making; uncertainties

1. Introduction

Carbon dioxide Capture and Storage (CCS) has put fossil fuels in a new lighting and made it possible to imagine and talk about a future with a maintained use of fossil fuels, while at the same time caring about the climate. Climate friendly coal is not necessarily considered an oxymoron today and has proponents even within the environmental movement. The framing of CCS in the policy making processes, mass media and within the research community has become increasingly positive [1, 2]. Initial optimism regarding new technologies is a generic phenomenon and may sometimes even be labeled as hypes [3]. Whether the future potential of CCS is hyped or not is still unclear, but the subject deserves attention since over-optimism may result in not only subsequent disappointment but more importantly: inappropriate regulatory adjustments and amendments (premature regulatory closure), sub optimal investments, oppositional public opinions and finally, in the case of CCS deployment, a deepened fossil fuel lock in. Experience from history of the development of nuclear power, as described by for example Anshelm [4] and Smil [5], may illustrate some of the above mentioned backlashes. The experience so far regarding the development of CCS shows potential signs of this over-optimism, e.g. the restructuring of Futurgen in the USA due to high costs, the cancellation of BP Rio Tinto Kwinana in Australia due to the lack of access to suitable storage sites, and Halten

* Corresponding author. Tel.: +46-8-7908285; fax: +46-8-7230858.

E-mail address: mrtn@kth.se.

c

Energy Procedia 1 (2009) 4673–4680

www.elsevier.com/locate/procedia

(3)

M. Bryngelsson, A. Hansson / Energy Procedia 00 (2008) 000–000

CO2 Project in Norway due to a feasibility study that claimed that the project, in spite of the technological feasibility, will not manage to become commercially viable [6]. CCS seems to be in a turbulent phase, regulatory frame works are being developed, expectations are rising, and pilot plants are constructed and attracting financial investments. However, CCS faces several significant scientific uncertainties [7, 8]. In order to deal with these uncertainties scenarios are constructed to support strategic decision making [9]. The purpose of a scenario is not to predict the future, rather to frame possible futures given a set of assumptions, prognoses or a specific storyline. In numerous previous studies the problems of energy systems modeling and scenario construction have been analyzed [10, 11, 12].

The purpose of this study is to i) analyze CCS scenarios, ii) discuss their function, and iii) suggest improvements. A few influential scenarios regarding the future role of CCS are studied and discussed. Key assumptions, results and recommendations are critically analyzed. The analysis is also briefly related to the ongoing policy process regarding CCS.

2. Methodology

In this study an analysis is performed on a small illustrative sample of scenario studies that include CCS. The studies are selected based on the criteria that they are either referred to in the Special Report on CCS (SRCCS) [7], the European Commission’s proposal for a directive on geological storage and its Impact Assessment [13, 14] or are essential reports for the European Technology Platform for Zero Emission Fossil Fuel Power Plants (ZEP) [15]. The following texts constitute the sample:

• [16] Edmonds, J.A. et al., The Role of Carbon Management Technologies in Addressing Atmospheric Stabilization of Greenhouse Gases, In: Proceedings of the 5th International Conference on Greenhouse Gas Control Technologies, Cairns, Australia, August 13-16, 2000.

• [17] McFarland, J.R. et al., Representing Energy Technologies in Top-down Economic Models Using Bottom-up Information, MIT Joint Program on the Science and Policy of Global Change, Cambridge, 2002.

• [18] Riahi, K et al., E. Towards fossil-based electricity systems with integrated CO2capture: implications of an illustrative long-term technology policy, In: Proceedings of the 7th International Conference on Greenhouse Gas Control Technologies, Vancouver, Canada, September, 5-9, 2004.

• [19] Stangeland, A Model for the CO2 Capture Potential, Int. J. of Greenhouse Gas Control No. 1 (2007) 418-429.

• [20] Capros, et al., Energy Systems Analysis of CCS technology – Primes Model Scenarios, Institute of Communication and Computer Systems, Athens, Greece, 2007.

As the sample only consists of five reports it should foremost be regarded as illustrative, even though the reports are very influential in policy and decision making contexts. The scenarios in the SRCCS related sample all include authors that were lead or coordinating lead authors in SRCCS. The reports are also repeatedly referenced to in the very same special report. Capros et al [20] is the main reference regarding scenarios in the EC draft directive [13] and Stangeland [19] is an important paper for the ZEP [15]. In the content analysis focus is on the following aspects: assumptions regarding fuel prices, learning rate, uncertainties and consistency regarding the framing of uncertainties.

3. The scenarios

In this section the major findings of the study are presented. First the SRCCS [7] related reports are discussed, closely followed by the EU related.

3.1. Special Report on CO2 Capture and Storage

The SRCCS is considered a milestone in the development of CCS. As the purpose of the Intergovernmental Panel on Climate Change is to provide decision makers with objective information on climate change, in combination with

(4)

their recommendations to pursue development of CCS, the report has become very influential. It is widely used in contexts where CCS is promoted. The future economic potential of CCS is analyzed in a literature review and is presented as being in the range of 60-95 % of the cumulative emission reductions during this century, assuming a stabilization of atmospheric CO2 at 450 ppm, even though many uncertainties regarding the performance of CCS’ components and aspects are emphasized. Among the major influential parameters are costs and availability of fossil fuels, technologies, rate of technological change and CO2 emissions. It is further stated that it is likely that the actual use of CCS will be lower than the estimated economical potential stemming from modeling. The conclusion contains a recommendation to look into knowledge gaps that affect the reliability of the models; there is a risk that the mitigation costs of CCS are underestimated due to models weaknesses. It is also emphasized that research is needed to assess CCS’ life cycle, include the impact of the recent increase in energy prices, and to employ a wider range of assumptions in order to better understand the sensitivity and robustness of the current scenarios. In the following section the SRCCS sample is analyzed and discussed [7].

In Edmonds et al. [16], a report referred to in the SRCCS to illustrate the future potential of CCS, two least-cost scenarios are analyzed, OGF (Oil and Gas Forever) and CBF (Coal Bridge to the Future). Both scenarios shows a massive deployment of CCS independently of stabilization targets. The scenarios also reflect some of the basic assumptions. In OGF the oil and gas resources are assumed to be “sufficiently abundant to maintain oil and gas

prices indefinitely” [16, p. 4]. In the CBF scenario the gas and oil prices increase but the coal resources are abundant

and the coal price is low. CCS is projected to be competitive with other mitigation options in the electricity sector by 2020. By the year 2095 the global mitigation cost reduction ranges between 68 and 81 % compared to scenarios in which CCS is not available. In spite of the statement that the experience of storing CO2 is very limited, it is assumed that “there is currently no reason to believe that any significant quantities of CO2 would be released to the

atmosphere” [16, p. 5]. The overall judgment of CCS is that “these technologies are potentially cost-competitive in a carbon constrained world.” In the abstract it is stated that “The value of these technologies [CCS] is robust regardless of whether the world’s economically recoverable oil and gas resources are eventually found to be large or small” [16, p. 4]. This statement witnesses of an assumption that there are no insurmountable barriers to the

development of CCS, which is in accordance with the analysis. However, important sensitivity analysis is missing. It is exemplified by the fact that the storage issue is ”black boxed” and that no life cycle analysis is performed. Furthermore, the costs of different fossil fuels are only altered between the two scenarios, and the coal resources are assumed to be abundant in both scenarios.

As in Edmonds [16] McFarland et al. [17] do not alter the coal price, only gas and oil prices. The coal price is low during the entire period until 2095. Riahi’s et al. [18] cost adjustment of coal up until 2100 is increasing from $1.2 per GJ in 2000 to $1.6 per GJ in 2050 and $2.6-2.7 per GJ in 2100. The relatively stable and low coal price is expected to favor the rapid growth of coal and CCS. The fossil fuel costs are explicitly described as dependent important variables in Riahi et al. [18] and are also emphasized in McFarland et al [17]. Still, the coal price is considered as being close to constant in the tree reports. The increasing gas price in McFarland’s scenario is presented as being one of the most important explanatory variables for the large scale deployment of coal fired CCS plants after 2050.

The novelty in the McFarland et al. study [17] is the inclusion of technological learning, which is assumed to follow a pattern similar to the penetration of nuclear power in the USA in the 1970’s and 1980’s. In the 1970’s the penetration share of nuclear power expanded by up to 45 % per year. It then decreased to 9 % per year in the 1980’s. The learning rate in Riahi et al. [18] is 12 %, a figure that “lie well within the ranges reported in the literature for

the diffusion of other successful technologies” [18, p. 3]. A penetration rate of 45 % is high, and the learning rate of

12 % also, considering that the average development of coal power related technologies in a historical perspective typically have been in a range between approximately 0.5 and 13 % [21] and the Integrated Gasification Combined Cycle which still after 30 years of development is not commercially mature [22].

McFarland et al. [17] present three different scenarios, 1) BAU, 2) CO2-tax ($25-200 per tonne), 3) and finally a scenario with a 550 ppm target for 2095. The 550 ppm target scenario shows the largest penetration for CCS.

(5)

Nuclear and renewables (except hydro) show a development close to zero in all scenarios. The only sensitivity analysis performed in this paper is capital vintaging. In spite of the statement that rising natural gas prices is the main explanatory factor for the replacement of CCS and natural gas with CCS and coal, no sensitivity analysis of fuel prices is performed. The study is said to have “reflected engineering estimates of feasible potential” [17, p. 22] witnessing of a perspective that the development takes place in an ideal world with no other than technical constrains. The argumentation continues: “Our underlying assessment of the cost of CCS technologies is based on

the technology as it exists today with only modest improvements” [17, p. 23]. This is potentially misleading since

there is no commercial electricity generation based on CCS and coal even today in 2008, six years after the publication of the study [6, 17].

Riahi et al. [18] analyze two scenarios, one with CCS and one without CCS. They are both based on IPCC’s A2 scenario from the special report on scenarios. A2 has a substantial increase in carbon emissions, low rate of technological change (delayed penetration of renewables), thus creating a higher need for CCS according to the authors. So, among the wide range of available SRES scenarios, only a scenario with assumptions that favors a significant CCS penetration was actively chosen. The study has very far-reaching conclusions: “The coal CCS

technology offers the most cost effective long-term source of low carbon emitting electricity, as the gas technologies are limited by gas resource availability reflected in high gas prices that make the technology non-competitive.” [18,

p. 23]. The largest uncertainty expressed is how climate policy is formulated and the improvement in CCS technology. Some other aspects are mentioned: “fossil fuel resources that directly determines future fuel prices, the

level of economic growth, energy efficiency improvement in the economy, and the other mitigation options available in the electric sector and in the economy in general”, but those aspects are neither included in a sensitivity analysis

nor do they have a negative effect on CCS deployment in the scenarios.

3.2. European Union

One activity under EU’s 7th_{Research Framework Programme concerns research, development and demonstration} of technologies to reduce the adverse environmental impact of fossil fuel, i.e. CCS. With these ambitions of the EU in mind, and in order to coordinate actions in the field, industrial stakeholders and the research community decided to create the Technology Platform for Zero Emission Fossil Fuel Power Plants (ZEP) in 2004. Their vision is to enable near-zero emissions from European fossil fuel power plants by 2020. In order to do that they work to drive down costs of CCS, they identify and remove obstacles as well as build public confidence in the technology. Currently EU works to promote the so-called EU Flagship Programme of 10-12 large-scale CCS projects across Europe by 2015. In its promotion of CCS, ZEP claims that: “With CCS, Europe can grow its economy, ensure a

secure energy supply …and meet its CO2 emission targets” [15, p. 2].

One of the main modeling studies that the ZEP refers to when making statements about the future role of CCS is Stangeland [19]. The calculations in the model are based on scenarios for future energy demand and CO2 emissions, mainly from IEA and the IPCC. The background and aim of the ZEP study is the following: “In most scenarios

from the literature, the potential for CCS is limited because of significant economical and political barriers that can delay the deployment of new technologies. The aim of the present work is therefore to calculate the full potential of CCS assuming that there are only minor political and economical barriers to wide implementation of CCS” [19, p.

419]. The two page list of assumptions that are made in order to remove the barriers for CCS includes: i) ignoring problems related to estimating CO2 storage potential by stating that the storage capacity is practically infinite, ii) assuming that 80 % of CO2 produced in the power sector within OECD is captured and stored by 2050. Stangeland claims 80 % capture is a conservative assumption since the capture rate is often higher (between 85 and 95 %) at individual power plants. However, when considering the whole life cycle, including accounting for all greenhouse gasses, an 80 % reduction is clearly not conservative, it is in the upper range [23]. Important aspects that are not mentioned include: the cost of the different scenario alternatives and externalities and environmental trade-offs associated with CCS such as increased fresh water consumption.

As Stangeland [19] mentions there are many uncertainties when it comes to the future role of CCS, depending on e.g. future policies and GHG emissions, future energy demand as well as penetration of renewables. Stangeland

(6)

briefly describes some of these uncertainties in the methodology section but avoids most of them when performing calculations by introducing a wide set of assumptions. Yet, these assumptions are not mentioned in the conclusions and abstract. The abstract reads: “The potential for CO2 capture has been calculated. In the EU, the accumulated

CO2 capture potential is 25 billion tones captured and stored by 2050. The global potential is 236 billion tonnes

CO2 captured and stored by 2050. This corresponds to 33 % reduction in global CO2 emissions in 2050 compared

to emissions today” [19, p. 428]. In this particular case when the assumptions are such an integral part of the results

they should be mentioned also in the conclusions and abstract. When looking how Stangelands results are used in promoting ZEP the point is made even clearer: “Indeed, if deployed to its full potential, CCS could reduce CO2

emissions in the EU by over 50 % by 2050. This includes not just the power sector, which alone accounts for around 30 %, but many other industry and transportation sectors as well” [19, p. 1]. When taken out of its context, the

expression “full potential” could be very misleading, making the reader think it has something to do with CCS’ competitiveness and cost-effectiveness, while it is CCS’ potential in an ideal world. The concept “full potential” shows similarities to McFarland’s et al. [17], expression mentioned above, “engineering potential”, which both seem hypothetical since they are restricted to be valid only in a world where uncertainties and constrains are excluded. The question is whether or not these modeled potentials are sometimes mistaken as realizable and desirable by decision makers.

The European Union draft directive on CO2 storage that was released in January of 2008 aims at enabling environmentally safe capture and geological storage of CO2 [13]. The grounds for the proposal are expressed in the following way: “Energy efficiency and renewables are in the long term the most sustainable solutions both for

security of supply and climate. However, we cannot reduce EU or world CO2 emissions by 50 % in 2050 if we do

not also use the possibility to capture CO2 from industrial installations and store it in geological formations“ [13, p.

2]. The explanation for this definite need for CCS stems from the urgency and scale of the climate change problem, and the notion that all mitigation options are needed including CCS.

In order to assess the implications of the deployment of CCS in Europe the draft directive was accompanied by an impact assessment (IA) [14]. In the IA the energy system was modeled by Capros et al. [20] using the so called PRIMES model, which is the main source for evaluating the different potential futures. The model “simulates the

European energy system and markets on a country-by-country basis and provides detailed results about energy balances, CO2 emissions, investment, energy technology penetration, prices and costs…” [20, p.2]. With the help of

this model several scenarios are constructed in order to show how CCS and the energy system might develop depending on different assumptions and policies up until the year 2030. The scenarios are divided into four options: i) no enabling policy for CCS at EU level, including no inclusion of CCS in the EU ETS, ii) Enable CCS under the EU ETS, iii) in addition to enabling under the ETS, impose an obligation to apply CCS from 2020 onwards and assess the impact on the potential positive externalities not captured by the carbon market, iv) in addition to enabling under the ETS, apply a subsidy so as to internalize the positive externalities not captured by the market [20].

One thing that may strike readers of Capros et al [20], and many other studies that are so-called computable general equilibrium models, is that it is very complex and the document on the internet called the “Energy Systems

Analysis of CCS technology - PRIMES MODEL SCENARIOS” is not very transparent. While a complex model may

impress some, it excludes many people from reading as well as understanding it, and there is little evidence that complex models do a better job of predicting or understanding the future than simple ones [5, 24]. Keepin & Wynne [24], who analyzed IIASA’s World Energy Study, a study that took 8 years to complete, concluded that the projections were very unstable and also based on informal guesswork. One thing they found was, among other things that small changes in the relative prices of technologies and resources could result in radically different outcomes.

Since relative prices of resources and technologies might give radically different output from models it is important that a wide range of prices are used as input, which would reflect the uncertain future of e.g. the fossil fuel prices. In Capros [20] fossil fuel prices are only mentioned when stating that the relative price of gas is high compared to that of coal, a statement similar to those in the scenarios described in section 3.1. However, in the EU

(7)

funded project ACCSEPT [8, p. 139] the prices used in the PRIMES model are presented, and it appears that even in the so-called “Soaring Prices Scenario” in Capros et al [20] use a drastically lower coal price in 2030 (€130 per toe) than the current European coal price in September 2008 (about $200 per toe according to the Swedish Coal Institute, 2008). Assuming low coal prices may increase the competitiveness of CCS from coal in comparison to e.g. renewables, thus potentially overestimating the future role of the technology.

External costs and life cycle effects of CCS are not mentioned in Capros et al [20]. The only external-like effect that is accounted for is employment, which in this particular case is a positive externality. Even if it is hard for the reader to understand how this factor is possible to account for, the result is that CCS is clearly beneficial from an employment point of view. According to the IA [14] the no-CCS scenario is the worst for European employment since 300 000 jobs would be lost due to closed mines and higher energy prices. However, since the higher energy prices in the no-CCS scenario depend on e.g. a low coal price and perhaps also a negligence of potential negative externalities the employment figures might uncertain.

The sensitivity analysis in Capros et al [20] is made up of two scenarios with i) a higher CO2 storage costs in new EU member states and ii) no nuclear phase out in Belgium and Germany. The choices of factors that are included in Capros’ sensitivity analysis seem quite arbitrary. Given the recommendations from the SRCCS [7], recent project failures, and the price surge of coal, steel and capital, one could have expected a sensitivity analysis on prices instead.

Table 1. Summary of important issues in the reports.

Some indicative patterns were found in the limited sample. All studies in the sample present the scenarios with major CCS implementation as the most desirable in terms of economical or abatement efficiency. No alternative scenarios were analyzed and in the abstracts of the scenario studies uncertainties are often given a marginalized role. Some important uncertainties are briefly presented inside the reports but are in several cases not included in the actual modeling. So, even though uncertainties are mentioned they often do not have an impact on the outcome, this also applies to the two studies written after the above-mentioned recommendations in SRCCS [7]. Several important aspects are not included (or only to a limited extent) even in the cases when sensitivity analyses are performed, e.g. coal costs, limitations regarding CO2 storage, CO2 seepage, LCA, political restrictions. There seems to be a discrepancy between the certainties of the conclusions, i.e. CCS successful implementation, and the uncertainties regarding CCS that are presented in the reports. In previous research by Hansson [2], similar discrepancy was found between experts’ spoken judgments about CCS and modeling results. Significant

2_{The issue is discussed in the conclusions and concluded as being poorly understood and in the need for further research.} 3_{High storage costs in new member states is one scenario. The marginal storage cost is increasing.}

4_{Alternative capital vintaging and malleability approaches are investigated.} 5_{The rate of produced CO}

2 that is captured and stored from different sectors and regions are varied in a sensivity analysis. Even though 18

different cases are included in the sensivity analysis the rate of CO2 captured and stored in different industry sectors do not vary more than 10

percentage points. Two different methods are used to calculate CCS potential. Depending on method the potential varies between 1.9-3.8Gt capture potential 2030.

6_{Storage costs, nuclear phase out}

Issue Report

Discussion on fossil fuel resources 16, 17 Low coal price assumption 16, 17, 18, 19, 20 Assumes perfect CO2 storage 16, 172, 18, 19, 203

LCA perspective

Sensitivity analysis 174_{, 19}5_{, 20}6

(8)

uncertainties were mentioned during interviews with the experts behind influential CCS scenario studies, but the uncertainties had in some cases limited impact on scenario results and subsequent conclusions [2].

4. Concluding discussion

Historical experience of energy forecasting shows that the accuracy of predictions has been poor. Utgikar and Scott [25] have analyzed a sample of energy experts’ predictions from the 1970’s regarding the development of emerging energy technologies. In retrospect the estimations in a significant number of cases seem spectacular. An illustrative example is the energy experts’ belief in e.g. a major breakthrough of fuel cells, breeder reactors and fusion power by the year 2000. To overcome some of the inherent problems of energy forecasting, scenario construction gained popularity in the 1980’s [12]. The purpose of scenarios is still debated. Here we divide scenarios into two broad categories: The first category presented by Armstrong [26] is that scenarios are useful in helping decision makers confront unpleasant or unexpected futures, which seems similar to Postma’s & Liebl’s [9] standpoint of incorporating “paradoxical trends” in the scenarios. This is not an easy task since “unpleasant scenarios” or forecasts are often are ignored [9, 11, 27]. The scenarios in our brief review were entirely pleasant from the viewpoint of a CCS proponent. However, “unpleasant” scenarios may also be of great value, for both proponents and opponents of CCS, because they may help decision makers prepare flexible strategies. The scenario process could in this case ease the use of information that contradicts the readers’ current world views. Scenarios that are in line with this category should include uncertainties and low-probability input data, especially since, as Postma & Liebl [9] state: in retrospect surprisingly often low probability scenarios mirror the actual outcome.

The second category of scenarios has the function of creating acceptance of a particular future. Previous research suggests that, the ease with which a scenario is imagined or presented as well as its plausibility, positively biases the beliefs that the suggested scenario will occur [12]. The scenarios help to boost the optimism regarding the development of a specific technology. Optimism is in many cases necessary in order to gain financial, political and societal support. According to Gartner’s [3] technology hype cycle the expectations of an emerging technology is often inflated with over-optimism and unrealistic expectations. In our review we find several examples of optimism, or even over-optimism, e. g. explicit assumptions on CCS being developed to its full technical potential and that CCS works flawlessly. The outcome is a massive deployment of CCS, in some studies these kind of scenarios are the only ones presented. Gartner [3] claims that failure to meet high expectations may backlash and divert attention and resources from the emerging technology. Recently, several backlashes for CCS have occurred, projects have exceeded their project budgets and coal prices have increased dramatically. More of these (un)expected events will probably occur considering the potential impacts of the excluded uncertainties in the CCS scenario literature. Society and decision makers have to be prepared, because it is likely that new problems may arise as we move towards full-scale deployment.

Using CCS scenarios as a tool for gaining acceptance for CCS seems to be prevalent in our sample. Therefore, scenarios that prepare society in case of “unpleasant” events are needed as well, some events may even be considered probable. Scenarios are great tools when preparing for unpleasant futures and such scenarios deserve more support. In agreement with Postma & Liebl [9] and SRCCS [7], we argue for a methodology where consequences of uncertainties are included in the scenario work and the importance of uncertainties is stressed in conclusions and results. Then the scenario could fill the purpose of an “early warning system”. When considering the experience of emerging energy technologies in the past, the following are examples of aspects that should be given more attention in modelling: under what circumstances (other than the absence of a carbon prices and regulatory frameworks) is CCS not viable, future coal price, assumptions regarding learning rates, leakage rates, environmental trade-offs and life cycle impacts of CCS. New scenarios that include these uncertainties and barriers are needed to balance the potential over-optimism, hence making the basis for decision making less shaky, or at least more nuanced.

(9)

Acknowledgements

We whish to thank the Swedish Energy Agency, Mistra’s Climate Policy Research Programme (Clipore) and the research group Technology, Values and Political Processes (TVOPP) at Linköping University.

References

[1] S. Mander, C. Gough, Media Framing of New Technology – the Case of Carbon Capture and Storage”, In: Proceedings of the 8th International Conference on Greenhouse Gas Control Technologies, Trondheim, Norway, June 19-22, 2006.

[2] A. Hansson, Return of Coal – Carbon dioxide capture and storage in science and politics, Linköping University, The Tema Institute, Technology and Social Change, 2008.

[3] A. Linden and J. Fenn, Understanding Gartner’s Hype Cycles, R-20-1971, Gartner Research, 2003.

[4] J. Anshelm, Mellan Frälsning och Domedag - Om Kärnkraftens Politiska Idéhistoria i Sverige 1945-1999, Symposium, Eslöv, Sweden, 2000. In Swedish.

[5] V. Smil, Perils of Long-Range Energy Forecasting - Reflections on Looking Far Ahead, Technological Forecasting and Social Change No. 3 (2000) 251-264.

[6] MIT, Carbon Dioxide Capture and Storage Projects, MIT, Cambridge, 2008. Available on http://sequestration.mit.edu/tools/projects/index.html

[7] IPCC, IPCC Special Report on Carbon Dioxide Capture and Storage, Prepared by Working Group III of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2005.

[8] DNV, J. Anderson, H. de Coninck, P. Curnow, T. Flach, D. Reiner, P. Richardson, S. Shackley, P. Upham, G. Sigurthorsson, The ACCSEPT Project: Multidisciplinary Analysis and Gap-filling Strategies, 2007. Available on http://www.accsept.org.

[9] T. J. B. M. Postma and F. Liebl, How to improve scenario analysis as a strategic tool? Technological Forecasting & Social Change ( 2005) 161-173.

[10] V. P. Utgikar, J.P. Scott, Energy forecasting: Predictions and analysis of causes of error, Energy Policy No. 17 (2006) 3087-3092. [11] T. Baumgartner, A. Midttun (eds.), The Politics of Energy Forecasting – A Comparative Study of Energy Forecasting in Western Europe and

North America, Clarendon Press, Oxford, 1987.

[12] W.L. Gregory and A. Duran, Scenarios and Acceptance of Forecasts, In J.S. Armstrong (ed.): Principles of Forecasting: A Handbook for Researchers and Practitioners, Springer, New York, 2001.

[13] EC, Proposal for a Directive of the European Parliament and of the Council on the geological storage of carbon dioxide, COM (2008) 18 Final, Brussels, January 23, 2008.

[14] EC, Impact Assessment, Accompanying Document to the Proposal for a Directive of the European Parliament and of the Council on the geological storage of carbon dioxide, COM (2008) XXX, Brussels, January 23, 2008.

[15] ZEP, CO2 Capture and Storage (CCS) Why it is essential to combat global warming. Brochure, May, 2008.

[16] J.A. Edmonds, P. Freund and J.J. Dooley, The Role of Carbon Management Technologies in Addressing Atmospheric Stabilization of Greenhouse Gases, In: Proceedings of the 5th International Conference on Greenhouse Gas Control Technologies, Cairns, Australia, August 13-16, 2000.

[17] J.R. McFarland, J. Reilly and H.J. Herzog, Representing Energy Technologies in Top-down Economic Models Using Bottom-up Information, MIT Joint Program on the Science and Policy of Global Change, Cambridge, 2002.

[18] K. Riahi, L. Barreto, S. Rao and Rubin, E. Towards fossil-based electricity systems with integrated CO2 capture: implications of an

illustrative long-term technology policy, In: Proceedings of the 5th International Conference on Greenhouse Gas Control Technologies, Cairns, Australia, August 13-16, 2000.

[19] A. Stangeland, A Model for the CO2 Capture Potential, Int. J. of Greenhouse Gas Control No. 1 (2007) 418-429.

[20] P. Capros, L. Mantzos, V. Papandreou, N. Tasios, A. Mantzaras, Energy Systems Analysis of CCS technology – Primes Model Scenarios, Institute of Communication and Computer Systems, Athens, Greece, 1997.

[21] T. Jamasb and J. Jamasb, Learning curves for energy technology and policy analysis: A critical assessment, University of Cambridge, 2007. [22] D. G. Ockwell, J. Watson, G. MacKerron, P. Pal and F. Yamin. Key Policy Considerations for facilitating low carbon transfer to developing

countries. Energy Policy 36 (2008) 4104-4115.

[23] J. Koornneef, T. van Keulen, A. Faaij and W. Turkenburg, Life Cycle Assessment of a Pulverized Coal Power Plant with Post-Combustion Capture, Transport and Storage of CO2, Int. J. of Greenhouse Gas Control No. 4 (2008) 448-467.

[24] B. Keepin and B. Wynne,The Roles of Models – What Can We Expect from Science? - A Study of the IIASA World Energy Model, In T. Baumartner and A. Midttun (eds.): The Politics of Energy Forcasting – A Comparative Study of Energy Forcasting in Western Europe and North America. Clarendon Press, Oxford, 1987.

[25] V. P. Utgikar, J.P. Scott, Energy forecasting: Predictions and analysis of causes of error, Energy Policy No. 17 (2006) 3087-3092. [26] J.S. Armstrong (ed.): Principles of Forecasting: A Handbook for Researchers and Practitioners, Springer, New York, 2001. [27] M.H. Bazerman, Climate Change as a Predictable Surprise, Climatic Change No. 1-2 (2006) 179-193.